This is an example of high-difficulty image processing using mathematical algorithms. In fact, the mathematical algorithm of image processing has developed to the point of breathtaking.
Scriptol lists several magical image processing algorithms, let's take a look.
An algorithm for generating vector graphs of pixel graphs
Early images in the digital age, with very low resolution. In particular, some video game pictures, zoomed in is a pixel block. The depixelizing algorithm enables low-resolution pixel graphs to be converted to high-quality vectors.
Coloring algorithm of black and white pictures
Let the old photo automatically turn into a color algorithm.
Third, the algorithm to eliminate shadows
An algorithm that removes the shadow of something on a photo without leaving a trace.
Iv. Algorithms for HDR photos
The so-called "HDR photo", is to enlarge the contrast between the light and dark, bright place becomes very bright, dark place becomes very dark, bright dark part of the details are very obvious.
There are many software implementations for HDR, and the G ' MIC is recommended here. It is a plugin for GIMP image editing software, all open source code.
Five, the elimination of debris algorithm
The so-called "eliminate clutter" is to draw an area on the photo and then fill it with the background automatically. Resynthesizer can do this and it's also a plugin for GIMP.
Six, the algorithm of automatic synthesis of photos
Based on a sketch, select the original photo and then compose it together to create a new photo. This is the scientific research of Tsinghua University.
Seven, beauty algorithm
An algorithm that automatically "beautify" the appearance.
The magic algorithm of image processing